 |
 Previous Article | Next Article 
The Journal of Neuroscience, January 1, 1999, 19(1):193-205
NMDA Receptor-Mediated Control of Presynaptic Calcium and
Neurotransmitter Release
Amanda J.
Cochilla and
Simon
Alford
Department of Physiology and Northwestern University Institute for
Neuroscience, Northwestern University Medical School, Chicago,
Illinois 60611
 |
ABSTRACT |
Before action potential-evoked Ca2+ transients,
basal presynaptic Ca2+ concentration may profoundly
affect the amplitude of subsequent neurotransmitter release.
Reticulospinal axons of the lamprey spinal cord receive glutamatergic
synaptic input. We have investigated the effect of this input on
presynaptic Ca2+ concentrations and evoked release
of neurotransmitter. Paired recordings were made between reticulospinal
axons and the neurons that make axo-axonic synapses onto those axons.
Both excitatory and inhibitory paired-cell responses were recorded in
the axons. Excitatory synaptic inputs were blocked by the AMPA
receptor antagonist 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX; 10 µM) and by the NMDA receptor antagonist
2-amino-5-phosphonopentanoate (AP-5; 50 µM).
Application of NMDA evoked an increase in presynaptic Ca2+ in reticulospinal axons. Extracellular
stimulation evoked Ca2+ transients in axons when
applied either directly over the axon or lateral to the axons.
Transients evoked by the two types of stimulation differed in magnitude
and sensitivity to AP-5. Simultaneous microelectrode recordings from
the axons during Ca2+ imaging revealed that
stimulation of synaptic inputs directed to the axons evoked
Ca2+ entry. By the use of paired-cell recordings
between reticulospinal axons and their postsynaptic targets, NMDA
receptor activation was shown to enhance evoked release of transmitter
from the axons that received axoaxonic inputs. When the synaptic input
to the axon was stimulated before eliciting an action potential in the axon, transmitter release from the axon was enhanced. We conclude that
NMDA receptor-mediated input to reticulospinal axons increases basal
Ca2+ within the axons and that this
Ca2+ is sufficient to enhance release from the axons.
Key words:
presynaptic calcium; glutamate receptor; NMDA receptor; non-NMDA receptor; transmitter release; presynaptic modulation
| |
INTRODUCTION |
Presynaptic modulation is a
widespread phenomenon in the CNS that occurs on many different time
frames, from milliseconds (Swandulla et al., 1991 ; Smith et al., 1993;
Tank et al., 1995) to days (Bliss and Lomo, 1973; Ekerot and Kano,
1985; Ito, 1989; Bliss and Collingridge, 1993). A leading model for
presynaptic facilitation of neurotransmitter release is that elevated
residual Ca2+ remains bound to release sites between
action potentials (Katz and Miledi, 1968; Kamiya and Zucker, 1994;
Bertram, 1997). Short-term enhancement of synaptic transmission
involving sustained presynaptic Ca2+ concentrations
may also be Na+ dependent (Mulkey and Zucker, 1992;
Delaney and Tank, 1994). Alternatively, at many synapses, including the
Xenopus neuromuscular junction (Fu and Huang, 1994) and
synapses in the rat amygdala (Huang et al., 1996), cerebellum (Sabatini
and Regehr, 1995), and hippocampus (Sabria et al., 1995), presynaptic
enhancement of release is mediated by increased presynaptic
Ca2+ conductance. Release of Ca2+
from intracellular stores has also been implicated in presynaptic facilitation in goldfish retina (Kobayashi et al., 1995) and lamprey spinal cord (Cochilla and Alford, 1998).
In the lamprey, reticulospinal axons make glutamatergic synapses with
spinal neurons. These synapses comprise mixed electrical and chemical
contacts with dendrites throughout the spinal cord (Rovainen, 1974;
Brodin et al., 1988). Consequently, recording from these axons is
electrotonically equivalent to recording within the terminal (Ringham,
1975). At the lamprey reticulospinal axon-motoneuron synapse,
facilitation of release occurs when metabotropic glutamate receptors
activate Ca2+-dependent Ca2+
release from internal stores (Cochilla and Alford, 1998). These reticulospinal axons also receive excitatory synaptic input (Cochilla and Alford, 1997) mediated by glutamate acting at AMPA and NMDA receptors.
Ionotropic receptors located at the presynaptic terminal may alter
transmission from the terminal by altering the Ca2+
concentration before action potential invasion. This mechanism of
enhancement of neurotransmitter release has been proposed for nicotinic
receptors in mossy fiber terminals of the hippocampus (Gray et al.,
1996). Ionotropic glutamate receptors, including kainate receptors
(Agrawal and Evans, 1986; Dev et al., 1996; Clarke et al., 1997; Kamiya
and Ozua, 1998), NMDA receptors (Berretta and Jones, 1996; Aoki et al.,
1997; Conti et al., 1997; Liu et al., 1997; Carlton et al., 1998; Chen
et al., 1998; Robert et al., 1998), NMDA-like receptors (Smirnova et
al., 1993), and AMPA receptors (Barnes et al., 1994; Farb et
al., 1995; Bureau and Mulle, 1998) have now been identified at
presynaptic terminals throughout the vertebrate neuroaxis. Their
actions in locations as diverse as the entorhinal cortex and the spinal
dorsal horn may be critical in damping neuronal excitability involved
in epilepsy and the regulation of pain pathways. These receptors might
also modulate release by altering presynaptic Ca2+
concentrations. This alteration could be either direct, by activating NMDA receptors or Ca2+-permeable AMPA or KA
receptors, or indirect, by depolarizing the terminal to activate
presynaptic voltage-operated Ca2+ channels (VOCCs).
We have investigated the role of presynaptic glutamate receptors in
modulating transmitter release from lamprey reticulospinal axons. We
conclude that axo-axonic activation of presynaptic NMDA receptors
enhances release from these axons by a
Ca2+-dependent mechanism.
Parts of this paper have been published previously (Holt et al.,
1996).
| |
MATERIALS AND METHODS |
The preparation. Lamprey ammoceotes
(Petromyzon marinus) were anesthetized with tricaine methyl
sulfonic acid (MS222; 100 mg/l), and sections of the spinal cord were
removed in accordance with institutional guidelines. After removal, the
tissue was maintained at 10°C in Ringer's solution containing
(concentration in mM): NaCl, 100; KCl, 2.1;
CaCl2, 2.6; MgCl2, 1.8; glucose,
4; and NaHCO3, 26 (adapted from Wickelgren, 1977).
The Ringer's solution was bubbled with 95% O2/5%
CO2 to a pH of 7.4. Whole-cell patch-clamp recordings were
achieved using the blind technique (Blanton et al., 1989; Alford and
Dubuc, 1993) with pipettes pulled to a DC tip resistance of 5-10 M
and filled with patch solution containing (concentration in
mM): cesium methane sulfonate, 102.5; NaCl, 1;
MgCl2, 1; EGTA, 5; and HEPES, 5. For sharp
microelectrode intracellular recordings, electrodes were pulled to a DC
tip resistance of 20-30 M when filled with 3 M
potassium methylsulfate. Drugs were administered by bath application.
For experiments in which only electrophysiological measurements were
made, the tissue was pinned ventral side up in a cooled Sylgard-floored
chamber to expose the ventromedial Reticulospinal axons for
intracellular and patch recording. The flow rate in the bath was ~1
µl/min, and the total volume of the chamber and access tubing was 1.1 ml.
Imaging experiments. The dextran amine-conjugated form of
the Ca2+ indicator dye Oregon green 488 BAPTA-1
(OGB-1; Molecular Probes, Eugene, OR) was used to label the axons
retrogradely. This labeling was accomplished by fitting a suction
electrode filled with dye over a cut end of the spinal cord immediately
after cutting the spinal cord. The tissue was then allowed to incubate
overnight, during which time the dye was transported throughout the
large reticulospinal axons in the spinal cord. Imaging experiments were performed with a confocal microscope (MRC600; Bio-Rad, Hercules, CA).
Images were obtained using a 20×, 0.75 numerical aperture Fluor
lens (Nikon). The 488 nm line of an argon ion laser was used as an
excitor source through a neutral density filter. Imaging of
evoked events was then performed by sampling two-dimensional arrays of
data (192 × 128 pixels) at 2 Hz. Images for data from wash-in of
agonist were sampled at 10 sec intervals. Higher speed microfluometric
recording was achieved by scanning repetitively the excitor laser over
the same line location at 500 Hz.
Analysis of the imaging data was performed on a Macintosh computer
using NIH Image software. NIH Image was used to calculate the
brightness value (range, 0-255 per 8 bit) for each pixel in the field
of view. For each individual axon of interest, the average brightness
value within the axon was measured, and background brightness (mode of
the entire field of view) was subtracted. The data were then normalized
to the baseline fluorescence to give  F/F
values, where 1 is the baseline value. Statistical analysis was
performed using either Microsoft Excel or IgorPro (WaveMetrics).
| |
RESULTS |
Reticulospinal axons have synaptic NMDA receptors
Previously, we have demonstrated that lamprey reticulospinal axons
receive glutamatergic synaptic input from other neurons in the spinal
cord (Cochilla and Alford, 1997). This input is mediated by NMDA and
AMPA receptors. Paired axonal recordings between reticulospinal axons
and their presynaptic inputs were made to demonstrate the nature of the
synaptic input directly (Fig. 1). An axon
was held under voltage clamp with a patch electrode, and a presynaptic
neuron was recorded under current clamp with a microelectrode. Both
inward and outward synaptic currents were recorded at holding
potentials of  70 mV in the voltage-clamped axons in response to
stimulation of different presynaptic axons. Responses recorded in one
axon in response to action potentials evoked in two different
presynaptic axons are demonstrated (Fig. 1A,B). For the excitatory input,
application of 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX; 10 µM) significantly reduced the amplitude of the evoked paired EPSC (reduced to 58.6 ± 6.7% of control;
p < 0.001; n = 7 pairs; Fig.
1A). In two of the pairs in which CNQX was applied the NMDA receptor antagonist 2-amino-5-phosphonopentanoate
(AP-5) was subsequently added to the superfusate. This further
reduced the amplitude of the remaining response (reduced to 11% of
control amplitude).
View larger version (18K):
[in this window]
[in a new window]
|
Figure 1.
Reticulospinal axons receive glutamatergic
synaptic inputs that act on AMPA and NMDA receptors. Paired recordings
were made between presynaptic neurons and postsynaptic axons.
Postsynaptic axons were held under voltage clamp with a patch
electrode, and presynaptic neurons were held under current clamp with a
microelectrode. A, A spike induced in a presynaptic
neuron by depolarizing current injection (lower trace)
evoked an EPSC in the postsynaptic axon (upper trace).
Application of CNQX (10 µM) reduced the
amplitude of the postsynaptic axonal EPSC. The remaining component of
the EPSC was abolished by the addition of AP-5 (100 µM; recording was in Ringer's solution with no added
Mg2+). B, Recordings from the same
postsynaptic axon as in A and a different presynaptic
neuron are shown. In this case, stimulation of the presynaptic neuron
evoked an IPSC in the postsynaptic axon. C, In another
postsynaptic axon, the spinal cord was stimulated to evoke a compound
EPSC (the tissue was superfused with Ringer's solution containing 2.6 mM MgCl2 and 10 µM bicuculline).
Addition of CNQX (10 µM) reduced the
amplitude of the response. D, The
CNQX-insensitive component of the EPSC was larger when
recorded at 50 mV than when recorded at 70 mV. E,
This component was reduced by the addition of AP-5 (50 µM) to the Ringer's solution (recorded at a holding
potential of 40 mV). F, Voltage-current plot of the
peak amplitude of the evoked EPSC between holding potentials of 40
and 70 mV in normal Ringer's solution is shown. Note that the
amplitude of the synaptic response increases with decreasing holding
potentials, indicating that Mg2+ block of the
channel is being relieved at depolarized potentials.
|
|
The AP-5-sensitive component of the EPSC that remains in CNQX was
investigated using recordings from 17 reticulospinal axons. Recordings
were made in the presence of bicuculline (10 µM) to block
GABAergic input to the axons. Stimulation of the spinal cord lateral to
the axon tracts (1 msec stimuli at <15 pA) evokes the excitatory
synaptic input to the axon (Cochilla and Alford, 1997). This response
is sensitive to CNQX (10 µM; n = 17 out
of 17; Fig. 1C). If the remaining component were NMDA
receptor-mediated, then the current-voltage relationship in the
presence of extracellular Mg2+ should have a
negative slope at membrane potentials negative to 40 mV, indicating
that Mg2+ block of the NMDA channel is relieved as
the membrane is depolarized. To test this hypothesis, we stepped the
membrane potential of the axon in 10 mV increments from 70 to 40
mV, and an EPSC was evoked at each holding potential (Fig.
1D). The amplitude of the EPSC increased as the
membrane potential was depolarized (n = 6 out of 11).
This region of negative slope in the voltage-current relationship
(Fig. 1F) indicates that the response is voltage dependent in a manner equated with NMDA receptor-mediated synaptic transmission in neuronal dendrites and somata. This voltage-dependent component of the response is reduced by the NMDA receptor antagonist AP-5 (50 µM; n = 4 out of 4; reduced to
55.5 ± 3.7% of control; Fig. 1E), and the
response was invariably enhanced after the removal of
Mg2+ from the superfusate (n = 7 out
of 7) (see also Cochilla and Alford, 1997). At holding potentials more
depolarized than 40 mV, a steady-state voltage-activated outward
current was recorded in the axons. This current was not blocked by
intracellular Cs+, intracellular QX-314 (1 mM), or intracellular 4-AP (1 mM). This current
impeded voltage clamp of the axons, obscured the synaptic response at
depolarized potentials, and prevented the acquisition of a
voltage-current relationship for the response at potentials more
depolarized than 40 mV. Nonetheless the sensitivity to AP-5 and the
negative slope of the current-voltage relationship between 70 and
40 mV indicate that this component of the EPSC is mediated by
glutamate acting at NMDA receptors on the reticulospinal axon.
Presynaptic NMDA receptors cause presynaptic
Ca2+ entry
Because NMDA receptors are permeable to Ca2+
(MacDermott et al., 1986), activation of these receptors in
axons has the potential to alter presynaptic Ca2+
levels within the axons. To test this hypothesis directly, we retrogradely labeled axons with the fluorescent Ca2+
indicator dye OGB-1, and fluorescence levels were monitored during NMDA
wash-in and wash-out (Fig. 2).
These experiments were performed in the presence of TTX (1 µM) to block Na+ action potentials.
All of the axons in the field of view (n = 7; Fig.
2B) increased in brightness during NMDA wash-in (500 µM; brightness increased by 30%). Brightness levels
decreased to baseline after wash-out of NMDA. The effect of wash-in of
NMDA was also tested for effect against depolarization of the axons.
Axons were recorded in the presence of TTX (1 µM) using
sharp microelectrodes containing cesium methylsulfate (3 M). All of the axons demonstrated a depolarization on
application of NMDA (500 µM; largest depolarization, 9 mV; mean, 5.7 ± 1.4 mV from a mean resting membrane potential of
76.3 ± 0.9 mV; n = 4; Fig. 2C). This
depolarization is insufficient to activate VOCCs in these axons because
the axons do not show any macroscopic Ca2+ channel
activation at membrane potentials negative to 40 mV (Cochilla and
Alford, 1998). We conclude that Ca2+ may enter
presynaptic elements through NMDA receptor ion channels.
View larger version (19K):
[in this window]
[in a new window]
|
Figure 2.
NMDA evokes a
Ca2+ transient. Axons were loaded with the
Ca2+ indicator dye OGB-1. NMDA (500 µM) was washed into the bath in the presence of TTX (1 µM), causing an increase in fluorescence in the axons.
A, Normalized fluorescence of one axon was recorded
before, during, and after wash-in of NMDA.
B, Normalized fluorescence of seven axons is shown
before, during, and after NMDA wash-in.
C, NMDA wash-in will depolarize the axons
by ~2-3 mV, which is not a substantial enough depolarization to
allow Ca2+ entry through VOCCs. A representative
axon recorded using a sharp microelectrode (recording made in 1 µM) TTX) is shown. Bath application of
NMDA led to a small depolarization (3 mV) in this
example, followed by an afterhyperpolarization.
|
|
Stimulus-evoked presynaptic Ca2+ transients have
an NMDA receptor-mediated component
Experiments were then performed to determine whether
stimulus-induced Ca2+ transients in the axons were
mediated directly by Ca2+-permeable NMDA receptors.
Axons were filled with OGB-1, and the spinal cord was stimulated
lateral to the axon tracts (1 msec stimuli at 50 Hz for 0.5 sec) to
activate synaptic inputs to the axons. This stimulation elicits a
Ca2+ transient in the axons that is sensitive to
external Mg2+. In the axon shown (Fig.
3A,B)
the amplitude of the Ca2+ transient in
Mg2+-free Ringer's solution was 176.2% of control.
The further addition of AP-5 (50 µM) to the
Mg2+-free superfusate decreased the response
amplitude to 31.5% of the amplitude in Mg2+-free
saline (Fig. 3B). For a group of axons, wash-out of
Mg2+ from the control Ringer's solution increased
the response to 166.4% of the amplitude in that control solution
(p < 0.05; n = 12; Fig.
3C). Addition of AP-5 (50 µM) to the
Mg2+-free Ringer's solution decreased the response
to 60.3% of the amplitude recorded in Mg2+-free
saline (p < 0.02; n = 6). The
reduction induced by AP-5 was reversible; wash-out of AP-5 into
Mg2+-free Ringer's solution increased the response
back to 152.4% of the control amplitude (n = 3; data
not shown). The Mg2+ sensitivity of the response may
be caused by Mg2+ acting at presynaptic
voltage-operated channels or by Mg2+-dependent block
of the NMDA receptor. However, the sensitivity of the response to the
application of AP-5 clearly indicates that axonal NMDA receptors
participate in the response of the axons to a train of stimuli applied
to the spinal cord.
View larger version (50K):
[in this window]
[in a new window]
|
Figure 3.
Ca2+ transients recorded in
axons are sensitive to Mg2+ and AP-5.
A, An axon was filled with OGB-1 dye, and we recorded
the response to a train of stimuli (50 Hz stimulation for 0.5 sec)
applied to the spinal cord. Left, The baseline
fluorescence. Middle, The peak fluorescence.
Right, The fluorescence 5 sec after stimulation
(stim.). The field is 53 × 80 µm in size.
B, The normalized brightness within the axon is graphed
at 0.5 sec intervals. The stimulus was delivered at the time indicated
by the horizontal bar. Responses in normal superfusate,
in Mg2+-free superfusate, after addition of
AP-5 (50 µM) to the
Mg2+-free superfusate, and after wash-out of
AP-5 are shown. The images in A were
taken at the times indicated in the graph (pre,
peak, and post). C, The
normalized brightness for a group of axons is shown. The stimulus was
delivered at the time indicated by the horizontal bar.
Responses in normal Ringer's solution (n = 12), in
Mg2+-free Ringer's solution (n = 12), after addition of AP-5 (50 µM;
n = 6) to the Mg2+-free
Ringer's solution, and after wash-out of AP-5
(n = 3) are shown. Graph data are mean ± SE. There is a significant difference in the peak brightness
between responses measured in control versus
Mg2+-free (p < 0.05) and
in Mg2+-free versus AP-5
(p < 0.02).
|
|
Axons show both spike-induced Ca2+ entry and
synaptically activated Ca2+ entry
During experiments involving single-stimulus locations alone, it
was not possible to determine whether extracellular stimulation lateral
to the axons evoked action potentials in the reticulospinal axons.
Consequently, stimulus-evoked Ca2+ entry into the
axons may have resulted either from the activation of excitatory
axo-axonic synapses (Cochilla and Alford, 1997) or from activation of
VOCCs. Experiments were performed to determine whether the
Ca2+ entry into the axon was caused by synaptic
activation of NMDA receptors at axo-axonic synapses. If so,
Ca2+ entry caused by a direct axonal stimulus should
differ from Ca2+ entry caused by an indirect
stimulus involving the activation of an axo-axonic synapse. Direct
stimulation of the axons will evoke action potentials in the axons and
cause Ca2+ entry through VOCCs, whereas indirect
stimulation of the spinal cord lateral to the axons should activate
neurons that synapse onto the axons and cause Ca2+
entry through synaptic receptors. To test this hypothesis, we used two
electrodes to stimulate the tissue both directly over the axon tracts
(direct stimulus) and lateral to the axon tracts (indirect stimulus;
see Fig. 4A for
schematic of recording setup). Axons were retrogradely labeled with the
Ca2+ indicator dye OGB-1, and the spinal cord was
stimulated to elicit Ca2+ transients in the axons.
These experiments were performed in Mg2+-free
Ringer's solution and in the presence of CNQX (10 µM).
Both direct and indirect stimulation elicited Ca2+
transients in the axon (Fig. 4B,C).
In nine axons tested, the amplitude of Ca2+
transients elicited by indirect stimulation was 42.5 ± 3.2% of those elicited by direct stimulation (p < 0.002; Figs. 4C,
5A). The indirectly activated
Ca2+ transient was decreased to 45.9 ± 3.8%
of the control amplitude when AP-5 was added to the superfusate (50 µM; p < 0.05; n = 4; stimulus at 50 Hz, 0.2 sec; Figs. 4C, 5B). The
amplitude reduction induced by AP-5 was reversible. On wash-out of
AP-5, the amplitude of the Ca2+ transient increased
to 95.3% of the amplitude recorded in control solution. In the same
four axons, the directly activated Ca2+ transient
was not sensitive to AP-5 (50 µM; AP-5 response amplitude was 98.7% of control amplitude; p = 0.49; Figs.
4C, 5C). Thus, direct stimulation evokes
Ca2+ entry into the axons in an AP-5-insensitive
manner, whereas indirect stimulation evokes Ca2+
entry in an AP-5-sensitive manner. Yet, with indirect stimulation there
is a response remaining in the presence of AP-5, indicating that at
least some of the indirectly activated Ca2+ entry is
through channels other than NMDA channels.
View larger version (46K):
[in this window]
[in a new window]
|
Figure 4.
Ca2+ transients
evoked by direct and indirect stimulation. A, Schematic
of recording setup showing stimulating electrodes
(stim) positioned to activate reticulospinal
axons directly and indirectly. Imaging was performed from the region
within the box. B, Images of axonal
Ca2+ recorded before (a,
d), during (b, e), and
after (c, f) stimulation
of the spinal cord. a-c, The response to
indirect stimulation. d-f, The response
to direct stimulation. Each panel is 105 × 70 µm
large. The inset shown above exploded
from b shows a pseudocolor representation
of the normalized fluorescence change during the stimulus. This data
compares b with the prestimulus condition in a
Ci, Normalized fluorescence from the
bottom-most axon shown in the field in
B. The response of the axon to indirect stimulation is
shown for images recorded in normal Ringer's solution and in the
presence of AP-5 (50 µM).
Cii, Normalized fluorescence from the same axon in
response to direct stimulation. Fluorescence from images recorded in
normal Ringer's solution and in the presence of AP-5
(50 µM) are shown. Di, In another
preparation, an investigation of the response of a single axon to one
indirect stimulus. A single line was repetitively
scanned along a length of the axon at 500 Hz. The spinal cord adjacent
to the axon was stimulated to evoke a Ca2+
transient. Dii, The graphs were generated by measuring
the fluorescence at each time point and normalizing this to the
prestimulus level to show the rise in axonal Ca2+.
The addition of AP-5 (50 µM) significantly
reduced the amplitude of the evoked transient. The response recovered
to control amplitude after washout of
AP-5.
|
|
View larger version (18K):
[in this window]
[in a new window]
|
Figure 5.
Pooled data comparing Ca2+
transients evoked by direct and indirect stimulation. A,
Normalized fluorescence recorded in nine axons in response to indirect
and direct stimulation. B, Normalized fluorescence from
four axons in response to indirect stimulation in normal Ringer's
solution or in the presence of AP-5 (50 µM). C, Normalized fluorescence from the
same four axons in response to direct stimulation recorded in normal
Ringer's solution or in the presence of AP-5 (50 µM). Note that the Ca2+ transient
recorded in response to indirect stimulation is sensitive to
AP-5, whereas the transient recorded in response to
direct stimulation is not. Data are expressed as the mean ± SE. SEs for B and C are
calculated relative to the control response.
|
|
It is possible to extract an approximate change in
Ca2+ concentration associated with the fluorescence
transients observed. Measurements from standard Ca2+
concentrations using the dextran conjugate of Oregon green 488 BAPTA-1
reveal a Kd of 170 nM and a
10-fold increase in fluorescence between 0 mM
Ca2+ bound to a saturating concentration of 40 µM
(Fmax/Fmin = 10). Retrogradely labeled axons were permeabilized with ionomycin, and the
resultant fluorescence was quenched with Mn2+ or
Co2+. The ratio of resting fluorescence to minimum
fluorescence
(Fr/Fmin) was 4.5 (M. Takahashi and S. Alford, unpublished
observation). It is possible to estimate the magnitude of the
resting Ca2+ concentration and the value of peak
Ca2+ using these parameters and the following
equation:
Given the ratios of
Fmax/Fmin = 10 and
Fr/Fmin = 4.5, we can calculate resting [Ca2+]
as:
or resting[Ca2+] is 108 nM.
Similarly the maximum ratio of peak fluorescence to resting
fluorescence (Fig. 4B, inset) is
2.0. Consequently peak [Ca2+] is close to
saturation of the dye with a ratio
Fp/Fmin of 9. This
reveals a peak [Ca2+] during stimulation of 1.3 µM, which is essentially saturating during the
application of 10 stimuli. It should be emphasized that these values
are approximate. The data for minimum and maximum fluorescence could
not be obtained from the axons in which the experiments were performed.
This calculation does, however, provide a justification that the NMDA
receptor-mediated elevation in Ca2+ is significant.
Each stimulus provides a presynaptic rise in [Ca2+] of 200 nM or more.
To investigate the time course of Ca2+ entry under
more physiological conditions, we used a second method to record the
indirect stimulation-evoked Ca2+ transient. A single
line of the image over a reticulospinal axon (Fig.
4Di) was chosen for repetitive scanning at 500 Hz.
The preparation was bathed in normal saline containing 1.8 mM MgCl2. Indirect stimulation was applied to
the spinal cord adjacent to the recorded axon. A single stimulus was
applied, and the resulting Ca2+ transient was
recorded (Fig. 4Di). AP-5 (50 µM) was
applied to the superfusate and caused a significant and
reversible reduction in the amplitude of the evoked response (amplitude
was reduced to 53.2 ± 10.0% of control; n = 3;
Fig. 4Dii). It was not possible to rule out that
indirect stimulation causes spiking of the imaged reticulospinal axons,
either by synaptic input raising the axonal membrane potential to
action potential threshold (Cochilla and Alford, 1997) or by current
spread from the stimulating electrode causing a direct stimulation of
the axons.
To address this question, we made electrophysiological recordings of
the membrane potential of the axons during these experiments (1) to
monitor when the axon was firing action potentials versus when it was
responding with EPSPs and (2) to stimulate selectively the axon in
isolation from other axons. After retrogradely labeling the preparation
with OGB-1, a microelectrode was used to record from an axon within the
spinal cord that had been retrogradely labeled with
Ca2+-sensitive dye (see Fig. 6A
for schematic of recording arrangement). To locate visually the axon
from which recordings were being made, we injected depolarizing current
into the axon to elicit action potentials (Fig.
6Bi). The action
potentials elicited a Ca2+ transient in the axon.
The only axon in the field of view that showed a
Ca2+ transient in response to the current injection
was the axon from which recordings were being made with the
microelectrode (Fig. 6C, Dii). Once the axon was
identified, the extracellular stimulating electrode was positioned, and
the stimulus strength was adjusted to elicit EPSPs in the axon without
evoking action potentials (mean peak EPSP amplitude, 4.2 ± 0.5 mV; reduced to 2.9 ± 0.6 mV in AP-5; n = 8; Fig.
6Bii,C, Di). To maximize the
NMDA receptor-mediated response and eliminate other ionotropic
glutamate receptor-mediated responses, these experiments were performed
in Mg2+-free superfusate and in the presence of CNQX
(10 µM). The amplitude of the Ca2+
transient elicited by EPSPs was 25.4 ± 4.01% of the amplitude of
the Ca2+ transient evoked by action potentials
(p < 0.005; n = 8). The EPSP-mediated Ca2+ transient was sensitive to AP-5.
The amplitude of the Ca2+ transient evoked by
extracellular stimulation in the presence of AP-5 (50 µM)
was reduced to 3.0 ± 1.9% of the control response (p < 0.05; n = 4; Fig.
6Di). The action potential-mediated
Ca2+ transient was not sensitive to AP-5 (50 µM) and was 105.8 ± 5.7% of the control response
(p = 0.43; n = 4; Fig.
6Dii). From these experiments we conclude that during
axoaxonic EPSPs, Ca2+ enters the presynaptic element
through NMDA channels in response to an excitatory synaptic input that
impinges directly on the axon.
View larger version (54K):
[in this window]
[in a new window]
|
Figure 6.
Ca2+ transients evoked by
presynaptic action potentials differ from those evoked by EPSPs.
A, Schematic of recording setup showing a microelectrode
recording from an axon and a stimulating electrode positioned to
activate reticulospinal axons indirectly, i.e., synaptically. Imaging
was performed from the region within the box.
B, Axonal responses recorded with the microelectrode.
The resting potential of the axon was approximately 70 mV. These
recordings were made simultaneously to the fluorescence recordings in
C and D. Stimulus artifacts have been
removed for clarity. Bi, Action potentials recorded in
the axon in response to intracellular current injection (2 msec current
injections at 50 Hz for 1.0 sec). Bii, EPSPs recorded in
the axon in response to extracellular stimulation adjacent to the axon
(1 msec shocks at 50 Hz for 1.0 sec). C, Images of
axonal Ca2+ recorded before (pre
stim.) and during (peak) stimulation.
Right, The normalized fluorescence change between the
pre stim. and peak conditions. Top
row, The response to intracellular stimulation seen in
Bi. Bottom row, The response to
extracellular stimulation seen in Bii. The
microelectrode recording is from the axon marked by the
arrowheads. This is the only axon that shows a response
to intracellular stimulation as is clear in the normalized image. Each
panel is 158 × 106 µm in size.
Di, Normalized fluorescence for the axon recorded by the
microelectrode during extracellular stimulation. The axonal
fluorescence in response to extracellular stimulation is shown for
images recorded before and after addition of AP-5 (50 µM). Dii, The normalized fluorescence
recorded from the same axon in response to intracellular stimulation
shown for images recorded before and after addition of
AP-5 (50 µM).
|
|
Activation of presynaptic NMDA receptors changes axonal
failure rate
NMDA receptor-mediated Ca2+ entry might have
the ability to alter transmitter release from reticulospinal axons at
their synapses onto motoneurons. However, it was not possible to test
directly the effect of NMDA application on the amplitude of
neurotransmitter release from those axons because the postsynaptic
targets also possess NMDA receptors and activation of these receptors
leads to complex activity in the spinal cord (Brodin et al., 1985). Synaptically activated inputs to the axons could provide a transient input to the axons leading to Ca2+ entry via NMDA
receptors. This Ca2+ transient can last for seconds
after the stimulus (see Fig. 3). We wished to test the hypothesis that
axo-axonic synaptically mediated Ca2+ entry into the
axons was able to modulate glutamate release from these axons. To test
this hypothesis it was necessary to activate the synaptic input to the
axon while monitoring the synaptic efficacy of the output synapses of
the axons. To perform these experiments we made paired recordings from
a presynaptic axon and a synaptically coupled postsynaptic motoneuron.
The axon was held under current clamp with a microelectrode. This
allowed us both to monitor the membrane potential during extracellular
stimulation to ensure that the axons did not spike and to generate
spikes in the axon directly by current injection. The motoneuron was
held under voltage clamp with a patch electrode. The recording
arrangement is illustrated schematically in Figure
7A.
View larger version (24K):
[in this window]
[in a new window]
|
Figure 7.
Activation of synaptic input to reticulospinal
axon facilitates transmission from reticulospinal axon to motoneuron. A
paired recording was made between a presynaptic reticulospinal axon and
a postsynaptic motoneuron. The axon was held under current clamp with a
microelectrode, and the motoneuron was held under voltage clamp with a
patch electrode. A, Schematic to describe the recording
arrangement. The prestimulus was provided via an extracellular
stimulating electrode placed lateral to the recorded pair.
B, The firing of an action potential by an axon in
response to a brief (2 msec) depolarizing current injection
(top). This evoked an EPSC in the postsynaptic
motoneuron (bottom). C, Recordings from
the same pair of neurons shown in B. Before current was
injected to elicit an action potential in the axon, the spinal cord was
stimulated to elicit an EPSP in the axon
(prestim). This stimulation also elicited EPSCs
in the postsynaptic neuron. After the prestimulus, current was injected
into the axon to elicit an action potential and a resulting EPSC in the
postsynaptic neuron. D, Enlargement of the postsynaptic
response to the presynaptic action potential before and after the
application of a prestimulus. The response is clearly enhanced by the
prestimulus. All traces are averages of four sequential
responses.
|
|
The presynaptic axon was stimulated by passing brief depolarizing
current pulses to evoke action potentials. EPSCs were then recorded in
the postsynaptic neuron (Fig. 7B). The extracellular stimulating electrode was then used to apply a prepulse stimulus to the
spinal cord 100 msec before the direct stimulation of the axon through
the recording microelectrode. The prepulse stimulus amplitude was
always set low enough to ensure that it did not evoke an action
potential in the presynaptic axon. The amplitude of postsynaptic EPSCs
evoked by direct stimulation of the axon in the absence of a
prestimulus was compared with the amplitude of EPSCs that were evoked
in the presence of an extracellular prestimulus. In seven pairs tested,
the amplitude of the postsynaptic EPSC was augmented when the
presynaptic prestimulus was given (amplitude increased to 133.1 ± 13.5% of control; p < 0.05).
It is reasonable to hypothesize that the enhancement of the synaptic
response was mediated by a presynaptic action of the prestimulus. It
was difficult, however, to determine this directly from the amplitude
of the postsynaptic response. Failure rates of chemical synaptic
transmission were, therefore, analyzed in two of the recorded pairs in
which the failure rate of the chemical component of the synaptic
response was >30%. Presynaptic action potentials elicited mixed
chemical and electrical EPSCs in the postsynaptic neurons (Fig.
8Ai, Aii).
The chemical component of the EPSCs showed variations in amplitude to
sequential stimuli; however, the amplitude of the electrical component
remained constant (Fig. 8Aii). Monitoring the
amplitude of the electrical component of the postsynaptic response
during stimulation served as a good measure of presynaptic action
potential propagation. If the electrical component was not recorded,
then the presynaptic action potential was not consistently propagating
to the synaptic contact between the axon and the motoneuron, and the
data were subsequently excluded from analysis. In the example shown in
Figure 8, the number of failures was decreased when the extracellular
prestimulus was activated (63% failure rate without the prestimulus;
38% failure rate with the prestimulus). The decrease in failure rate
was blocked by AP-5 (50 µM; 54% failure rate with the
prestimulus in AP-5).
View larger version (28K):
[in this window]
[in a new window]
|
Figure 8.
Activation of synaptic input to
reticulospinal axon facilitates transmission from reticulospinal axon
to motoneuron. A, A paired recording made between a
presynaptic reticulospinal axon and a postsynaptic motoneuron (as in
Fig. 7). The axon was held under current clamp with a microelectrode,
and the motoneuron was held under voltage clamp with a patch electrode.
Ai, Action potential evoked in the axon in response to
current injection (top). The motoneuron responds to the
presynaptic action potential with a mixed chemical and electrical EPSC
(bottom). Aii, Enlargement of the
responses of the postsynaptic neuron to the presynaptic action
potentials. Individual traces are shown in
gray, and the average of the four traces
is shown in black. There is a failure rate of 63% for
the chemical component. Bi, Recordings from the same
pair of neurons shown in A. Before current was injected
to elicit an action potential in the axon (intracellular stimulus), the
spinal cord was stimulated to elicit an EPSP in the axon (extracellular
prestimulus). This stimulation also elicited EPSCs in the postsynaptic
neuron. After the prestimulus, current was injected into the axon to
elicit an action potential in the axon and a resulting EPSC in the
postsynaptic neuron as described for Figure 7. Bii,
Enlargement of the responses of the postsynaptic neuron to the
presynaptic action potentials. The failure rate of the EPSCs elicited
after the prestimulus is now 38%. Ci, The same protocol
described in Bi performed in the presence of
AP-5 (50 µM). Cii,
Enlargement of the responses of the postsynaptic neuron to the
presynaptic action potentials. The failure rate of the EPSCs elicited
after the prestimulus in the presence of AP-5 is
increased to 54%.
|
|
The data for this cell are summarized in Figure
9. The panels follow the same
sequence as those of Figure 8. Amplitude histograms are displayed
showing the marked reduction in failures of chemical transmission after
the application of a prestimulus. The amplitude histogram of the
electrical component of the synaptic response demonstrates that the
presynaptic action potential always invaded the terminal. In the two
axons analyzed for failures, an electrical component of the synaptic
response was always present. The reduction of failure frequency by the
prestimulus in an AP-5-dependent manner is consistent with the
presynaptic activation of NMDA receptors that enhance the probability
of neurotransmitter release.
View larger version (28K):
[in this window]
[in a new window]
|
Figure 9.
Amplitude histograms of paired synaptic responses
between a reticulospinal axon and a postsynaptic neuron. This data are
from the pair of neurons illustrated in Figure 8. A, The
EPSC amplitudes recorded after stimulation of only the presynaptic
reticulospinal neuron. Left, A histogram of all event
amplitudes. Those events marked in white were considered
failures. Events marked gray were amplitudes of
individual electrical events, in which the amplitude histogram showed a
similar distribution to the recording noise. The black
events were identified as chemical events. No event was included
(either electrical or chemical) if there was no preceding electrical
event. Right, Comparison of the rate of failures with
the existence of a chemical response regardless of the amplitude of the
chemical response. B, The EPSC amplitudes recorded on
stimulation of the reticulospinal axon 100 msec after the application
of a prepulse to the spinal cord. The panels are as
described for A. The application of a prepulse markedly
reduced the number of recorded failures. Note that the mean amplitude
of the electrical component was slightly reduced. This effect reflects
the synaptic drive from the prepulse on the postsynaptic neuron.
C, EPSC amplitudes recorded similarly to
B but after the addition of AP-5 (100 µM).
|
|
| |
DISCUSSION |
Whole-cell patch recordings in giant axons of the lamprey have
revealed individual synaptic inputs to the axons. Functionally, such
recordings are electrically equivalent to recording from the
presynaptic terminal. These axons possess no arborization and are known
to form en passant glutamatergic synapses with motoneurons and
interneurons along the entire length of the spinal cord (Rovainen, 1974; Shupliakov et al., 1992). Stimulation of the spinal cord lateral
to the ventromedial tracts where the reticulospinal axons are located
evokes a depolarization in the axons that is mediated by excitatory
synapses impinging on the axon (see Fig. 1). The depolarization is
mediated by glutamate acting at both AMPA and NMDA receptors (Cochilla
and Alford, 1997). It has become clear that ionotropic glutamate
receptors are present in mammalian presynaptic terminals (Agrawal and
Evans, 1986; Barnes et al., 1994; Farb et al., 1995; Smirnova et al.,
1993; Berretta and Jones, 1996; Dev et al., 1996; Aoki et al.,
1997; Clarke et al., 1997; Conti et al., 1997; Liu et al., 1997). Yet,
very little is understood of the effect of activation of these
receptors on release of transmitter. Our goal was to determine whether
Ca2+ entered the axons through the NMDA receptors
and, if so, whether the Ca2+ entry could alter
neurotransmitter release from the axon.
After wash-in of NMDA into the preparation, a Ca2+
transient is visible in the axons (see Fig. 2). This
Ca2+ entry is not attributable to
Ca2+ entry through voltage-operated
Ca2+ channels. Application of NMDA to the
preparation after blockade of Na+ channels with TTX
does not cause significant depolarization of the axons (see Fig. 3)
(Cochilla and Alford, 1997). We have shown previously (Cochilla and
Alford, 1998) that these axons do not possess any low voltage-activated
Ca2+ channels; voltage-operated
Ca2+ channels are not activated until the membrane
is stepped to potentials more positive than 40 mV. Because the
superfusion of NMDA will only depolarize the membrane by ~5 mV and
because there are no low voltage-activated voltage-operated
Ca2+ channels on these axons, we can conclude that
Ca2+ transients elicited by NMDA wash-in in the
presence of TTX are attributable to Ca2+ entry
through NMDA channels located on the axonal membrane.
Ca2+ entry into the axons can also be evoked by
stimulating the spinal cord lateral to the axon tracts (see Fig. 3).
The sensitivity of this transient to both Mg2+ and
AP-5 is indicative of NMDA receptor-mediated responses (Mayer et al.,
1984; Nowak et al., 1984). Ca2+ transients are also
initiated by direct axonal stimulation (see Fig. 4). However, these
Ca2+ transients are qualitatively different from the
transients evoked by indirect stimulation. Transients evoked by
indirect stimulation are sensitive to AP-5 and are smaller in amplitude
than Ca2+ transients evoked by direct axonal
stimulation. Direct electrophysiological recordings from axons during
Ca2+-imaging experiments allowed us to show
definitively that Ca2+ enters the presynaptic
element during EPSPs evoked in the axon that are subthreshold to the
activation of voltage-operated Ca2+ channels (see
Fig. 6). Similar to Ca2+ entry mediated by
application of NMDA to the preparation, the depolarizations activated
by synaptic inputs onto the axons (~5 mV from a resting membrane
potential of 70 mV; see Fig. 6Bi) are not large
enough to activate VOCCs.
The location of the NMDA receptor-mediated Ca2+
transient in reticulospinal axons is important in the determination of
whether the elevated Ca2+ concentration will alter
evoked-transmitter release from the axons. For the NMDA
receptor-dependent axonal Ca2+ transients
demonstrated in the reticulospinal neurons to alter evoked-transmitter
release from the axons, it is clear that the elevated
Ca2+ must be present at or near the presynaptic
specializations of the reticulospinal axons. Reticulospinal axons of
the lamprey spinal cord show numerous en passant synaptic connections
at all segmental levels in the spinal cord (Rovainen, 1974), such that motoneuronal dendrites surround the axons (see Fig. 10).
To illustrate the relationship between synaptically evoked presynaptic
Ca2+ entry, the presynaptic axon, and the location
of the output synapses from that axon, we made paired-cell recordings
to label and image reticulospinal axons and their postsynaptic target
neurons. The recording arrangement is shown in the schematic of Figure
10A. To affect
release from the axon, Ca2+ entering the axon
through synaptic NMDA receptors must be within close proximity to the
release sites of the axon. The hypothesis is illustrated in the
schematic in Figure 10B. That this scheme is possible
is illustrated by the reconstruction of the reticulospinal axon and its
postsynaptic target in Figure 10C. The
Ca2+ transients illustrated in Figures 4 and 6
clearly encompass a volume of axon that would contain numerous
presynaptic sites (for the axon shown in Fig. 4, ~75 µm of axon is
affected by the stimulus). Additionally, the Ca2+
transients recorded in response to action potential initiation in the
axon and to synaptic stimulation of inputs to the axon show
considerable overlap (see Fig. 6).
View larger version (44K):
[in this window]
[in a new window]
|
Figure 10.
Model explaining the mechanism of presynaptic
NMDA receptor enhancement of transmitter release. A,
Schematic demonstrating the relationship between reticulospinal
neurons and their postsynaptic targets. The electrodes shown
represent the approximate placement of electrodes used to fill the
neurons for the image in C. B, The
axoaxonic input to the reticulospinal neuron activating ionotropic
glutamate receptors, including NMDA receptors, on the presynaptic
reticulospinal axon (blue). This causes a wide diffusion
of Ca2+ in the axon (red). The extent
of the Ca2+ signal can be >500 µm along the axon
(as demonstrated in Fig. 6). This Ca2+ transient
will alter the basal Ca2+ concentration at numerous
synaptic terminals. In the schematic, the region of elevated calcium
encompasses glutamatergic vesicles and Ca2+
channels, and there is an enhancement of transmission onto postsynaptic
neurons. C, Three-dimensional reconstruction of a
synaptically coupled reticulospinal axon (dark vertical
structure) and postsynaptic premotor interneuron. The pair was
identified electrophysiologically using paired-microelectrode recording
between the two cells as shown in the schematic in A.
Both cells were filled with fluorescent dye (Lucifer yellow) by
pressure ejection through the microelectrodes. Images were obtained
using a confocal microscope. The boxed region of the
top is enlarged and rotated 90° on the bottom
left, to view along the long axis of the axon. On the
bottom right, two individual synaptic contacts
(boxed region of the bottom left) are
shown rotated at high magnification for improved visualization. This
image demonstrates the relationship between the structure of the axons
and the location of the en passant synaptic terminals. Each axon makes
many overlapping synaptic contacts similar to the contact shown. Thus,
an axoaxonically evoked Ca2+ rise that diffuses
hundreds of micrometers along the axon will affect
Ca2+ concentrations at numerous presynaptic
terminals in the reticulospinal axon.
|
|
Activation of excitatory axo-axonic synapses onto reticulospinal axons
enhances release from the axons. The mixed electrical and chemical
synapse between reticulospinal axons and motoneurons provides a
built-in control for determining whether the action potential has
propagated to the synapse. The electrical component of the response was
not blocked by previous activation of the axoaxonic synaptic input.
Neither did application of AP-5 affect the electrical component.
However, the chemical component was markedly enhanced by previous
activation of the extracellular prestimulus and subsequently depressed
by application of AP-5 (see Fig. 7). This enhancement is a presynaptic
phenomenon. Failure of the axon to release neurotransmitter causes
failures in the chemical component of a synaptic response. The failure
rate of the reticulospinal axon was reduced when the axoaxonic synapse impinging on the axon was stimulated (see Figs. 8, 9). When NMDA receptors were blocked by AP-5, the depression of the failure rate was blocked.
A number of mechanisms may account for the stimulus- and NMDA
receptor-dependent enhancement seen in this study. (1) Evoked axonal
NMDA receptor activation will depolarize the presynaptic axon. This may
lead to a greater activation of Ca2+ channels when
action potentials are evoked in the presynaptic axon. (2)
Alternatively, activation of presynaptic NMDA receptors may lead to a
presynaptic second messenger cascade after Ca2+
entry via these NMDA receptors. (3) Finally, this effect may be caused
by Ca2+ entry into the presynaptic element through
synaptic NMDA channels acting at the presynaptic release sites.
We propose that the simplest explanation is found in the third
mechanism. Activation of the axo-axonic synapse elicits EPSPs in the
axon that stimulate NMDA receptors and cause
Ca2+ entry into the presynaptic element. After only
one stimulus, axonal Ca2+ levels had not returned to
baseline levels after 1 sec (see Fig. 4). Thus, a localized area of
elevated baseline Ca2+ would exist within the axon
for seconds after activation of the synaptic input to the axon. Any
output synapse within that localized area would be affected by the
elevated Ca2+ that reaches at least hundreds of
nanomolar above resting levels with just one stimulus. Activation of
synaptic inputs to these axons thereby facilitates release from the
axon. Ca2+ enters the presynaptic element through
synaptic NMDA receptors located on the axon and diffuses to nearby
output synapses where it enhances neurotransmitter release from those
synapses. This enhancement occurs in a manner similar to one mechanism
that is proposed to underlie paired-pulse facilitation (Katz and
Miledi, 1968; Kamiya and Zucker, 1994; Bertram, 1997). If the axon
fires an action potential within a short time after receiving the
synaptic input or if the synaptic input drives the axon to fire an
action potential (Cochilla and Alford, 1997), neurotransmitter release from the axon would be enhanced at output synapses near the input synapse. In principle, the same mechanism of presynaptic
Ca2+-induced facilitation of transmitter release
would occur if the NMDA receptors were also positioned as autoreceptors
at the reticulospinal synaptic terminal. We are not, at this time, able
to test this hypothesis directly. However, NMDA receptors thus far
identified in mammalian synaptic terminals have been proposed to be autoreceptors.
| |
FOOTNOTES |
Received July 30, 1998; revised Oct. 9, 1998; accepted Oct. 15, 1998.
This work was supported by National Institute of Neurological Disorders
and Stroke Grants NS31713 and NS32114. We would like to thank Drs.
N. T. Slater and N. E. Schwartz for help in this study and
comments on this manuscript. We would also like to thank J. Hsu for his
help in reconstructing confocal data sets.
Correspondence should be addressed to Dr. Simon Alford, Department of
Physiology, Northwestern University Medical School, 303 East Chicago
Avenue, Chicago, IL 60611.
Dr. Cochilla's present address: Department of Physiology and
Biophysics, University of Colorado Medical School, 4200 East Ninth
Avenue Box C240, Denver, CO 80262.
| |
REFERENCES |
-
Agrawal SG,
Evans RH
(1986)
The primary afferent depolarizing action of kainate in the rat.
Br J Pharmacol
87:345-355[Web of Science][Medline].
-
Alford S,
Dubuc R
(1993)
Glutamate metabotropic receptor-mediated depression of synaptic inputs to lamprey reticulospinal neurons.
Brain Res
605:175-180[Web of Science][Medline].
-
Aoki C,
Rhee J,
Lubin M,
Dawson TM
(1997)
NMDA-R1 subunit of the cerebral cortex co-localizes with neuronal nitric oxide synthase at pre- and postsynaptic sites and in spines.
Brain Res
750:25-40[Web of Science][Medline].
-
Barnes JM,
Dev KK,
Henley JM
(1994)
Cyclothiazide unmasks AMPA-evoked stimulation of [3H]-L-glutamate release from rat hippocampal synaptosomes.
Br J Pharmacol
113:339-341[Web of Science][Medline].
-
Berretta N,
Jones RS
(1996)
Tonic facilitation of glutamate release by presynaptic N-methyl-D-aspartate autoreceptors in the entorhinal cortex.
Neuroscience
75:339-344[Web of Science][Medline].
-
Bertram R
(1997)
A simple model of transmitter release and facilitation.
Neural Comput
9:515-523[Web of Science][Medline].
-
Blanton MG,
Lo Turco JJ,
Kriegstein AR
(1989)
Whole cell recording from neurons in slices of reptilian and mammalian cerebral cortex.
J Neurosci Methods
30:203-210[Web of Science][Medline].
-
Bliss TVP,
Collingridge GL
(1993)
A synaptic model of memory: long-term potentiation in the hippocampus.
Nature
361:31-39[Medline].
-
Bliss TVP,
Lømo T
(1973)
Long-lasting potentiation of synaptic transmission in the dentate area of the anaesthetised rabbit.
J Physiol (Lond)
232:331-356[Abstract/Free Full Text].
-
Brodin L,
Grillner S,
Rovainen CM
(1985)
N-methyl-D-aspartate (NMDA), kainate and quisqualate receptors and the generation of fictive locomotion in the lamprey spinal cord.
Brain Res
325:302-306[Web of Science][Medline].
-
Brodin L,
Grillner S,
Dubuc R,
Ohta Y,
Kasicki S,
Hokfelt T
(1988)
Reticulospinal neurones in lamprey: transmitters, synaptic interactions and their role during locomotion.
Arch Ital Biol
126:317-345[Web of Science][Medline].
-
Bureau I,
Mulle C
(1998)
Potentiation of GABAergic synaptic transmission by AMPA receptors in mouse cerebellar stellate cells: changes during development.
J Physiol (Lond)
509:817-831[Abstract/Free Full Text].
-
Carlton SM,
Chung K,
Ding Z,
Coggeshall RE
(1998)
Glutamate receptors on postganglionic sympathetic axons.
Neuroscience
83:601-605[Web of Science][Medline].
-
Chen YH,
Wu ML,
Fu WM
(1998)
Regulation of presynaptic NMDA responses by external and intracellular pH changes at developing neuromuscular synapses.
J Neurosci
18:2982-2990[Abstract/Free Full Text].
-
Clarke VRJ,
Ballyk BA,
Hoo KH,
Mandelzys A,
Pellizzari A,
Bath CP,
Thomas J,
Sharpe EF,
Davies CH,
Ornstein PL,
Schoepp DD,
Kamboj RK,
Collingridge GL,
Lodge D,
Bleakman D
(1997)
A hippocampal GluR5 kainate receptor regulating inhibitory synaptic transmission.
Nature
389:599-603[Medline].
-
Cochilla AJ,
Alford S
(1997)
Glutamate receptor-mediated synaptic excitation in axons of the lamprey.
J Physiol (Lond)
499:443-457[Abstract/Free Full Text].
-
Cochilla AJ,
Alford S
(1998)
Metabotropic glutamate receptor-mediated control of neurotransmitter release.
Neuron
20:1007-1016[Web of Science][Medline].
-
Conti F,
Minelli A,
Debiasi S,
Melone M
(1997)
Neuronal and glial localization of NMDA receptors in the cerebral cortex.
Mol Neurobiol
14:1-18[Web of Science][Medline].
-
Delaney KR,
Tank DW
(1994)
A quantitative measurement of the dependence of short-term synaptic enhancement on presynaptic residual calcium.
J Neurosci
14:5885-5902[Abstract].
-
Dev KK,
Barnes JM,
Collingridge GL,
Henley JM
(1996)
Regulation of glutamate release by presynaptic kainate receptors in the hippocampus.
Nature
379:78-81[Medline].
-
Ekerot C-F,
Kano M
(1985)
Long-term depression of parallel fibre synapses following stimulation of climbing fibres.
Brain Res
342:357-360[Web of Science][Medline].
-
Farb CR,
Aoki C,
LeDoux JE
(1995)
Differential localization of NMDA and AMPA receptor subunits in the lateral and basal nuclei of the amygdala: a light and electron microscopic study.
J Comp Neurol
362:86-108[Web of Science][Medline].
-
Fu W-M,
Huang FL
(1994)
L-type Ca2+ channel is involved in the regulation of spontaneous transmitter release at developing neuromuscular synapses.
Neuroscience
58:131-140[Web of Science][Medline].
-
Gray R,
Rajan AS,
Radcliffe KA,
Yakehiro M,
Dani JA
(1996)
Hippocampal synaptic transmission enhanced by low concentrations of nicotine.
Nature
383:713-716[Medline].
-
Holt AJ,
Schwartz NE,
Alford S
(1996)
Presynaptic NMDA receptors control presynaptic calcium.
Soc Neurosci Abstr
22:494.
-
Huang CC,
Hsu KS,
Gean PW
(1996)
Isoproterenol potentiates synaptic transmission primarily by enhancing presynaptic calcium influx via P- and/or Q-type calcium channels in the rat amygdala.
J Neurosci
16:1026-1033[Abstract/Free Full Text].
-
Ito M
(1989)
Long-term depression.
Annu Rev Neurosci
12:85-102[Web of Science][Medline].
-
Kamiya H,
Ozua S
(1998)
Kainate receptor-mediated inhibition of presynaptic Ca2+ influx and EPSPs in area CA1 of the rat hippocampus.
J Physiol (Lond)
509:817-831.
-
Kamiya H,
Zucker RS
(1994)
Residual Ca2+ and short-term synaptic plasticity.
Nature
371:603-606[Medline].
-
Katz B,
Miledi R
(1968)
The role of calcium in neuromuscular facilitation.
J Physiol (Lond)
195:481-492[Abstract/Free Full Text].
-
Kobayashi K,
Sakaba T,
Tachibana M
(1995)
Potentiation of Ca2+ transients in the presynaptic terminals of goldfish retinal bipolar cells.
J Physiol (Lond)
482:7-13[Abstract/Free Full Text].
-
Liu H,
Mantyh PW,
Basbaum AI
(1997)
NMDA-receptor regulation of substance P release from primary afferent nociceptors.
Nature
386:721-724[Medline].
-
MacDermott AB,
Mayer ML,
Westbrook GL,
Smith SJ,
Barker JL
(1986)
NMDA-receptor activation increases cytoplasmic calcium concentration in cultured spinal cord neurones.
Nature
321:519-522[Medline].
-
Mayer ML,
Westbrook GL,
Guthrie PB
(1984)
Voltage-dependent block by Mg2+ of NMDA responses in spinal cord neurones.
Nature
309:261-263[Medline].
-
Mulkey RM,
Zucker RS
(1992)
Posttetanic potentiation at the crayfish neuromuscular junction is dependent on both intracellular calcium and sodium ion accumulation.
J Neurosci
12:4327-4336[Abstract].
-
Nowak L,
Bregetovski P,
Ascher P,
Herbet A,
Prochiautz A
(1984)
Magnesium gates glutamate activated channels in mouse central neurons.
Nature
307:462-465[Medline].
-
Ringham GL
(1975)
Localization and electrical characteristics of a giant synapse in the spinal cord of the lamprey.
J Physiol (Lond)
251:395-407[Abstract/Free Full Text].
-
Robert A,
Black JA,
Waxman SG
(1998)
Endogenous NMDA-receptor activation regulates glutamate release in cultured spinal neurons.
J Neurophysiol
80:196-208[Abstract/Free Full Text].
-
Rovainen CM
(1974)
Synaptic interactions of identified nerve cells in the spinal cord of the sea lamprey.
J Comp Neurol
154:189-206[Web of Science][Medline].
-
Sabatini BL,
Regehr WG
(1995)
Detecting changes in calcium influx which contribute to synaptic modulation in mammalian brain slice.
Neuropharmacology
34:1453-1467[Web of Science][Medline].
-
Sabria J,
Pastor C,
Clos MV,
Garcia A,
Badia A
(1995)
Involvement of different types of voltage-sensitive calcium channels in the presynaptic regulation of noradrenaline release in rat brain cortex and hippocampus.
J Neurochem
64:2567-2571[Web of Science][Medline].
-
Shupliakov O,
Brodin L,
Cullheim S,
Ottersen OP,
Storm-Mathisen J
(1992)
Immunogold quantification of glutamate in two types of excitatory synapse with different firing patterns.
J Neurosci
12:3789-3803[Abstract].
-
Smirnova T,
Stinnakre J,
Mallet J
(1993)
Characterization of a presynaptic glutamate receptor.
Science
262:430-433[Abstract/Free Full Text].
-
Smith SJ,
Buchanan J,
Osses LR,
Charlton MP,
Augustine GJ
(1993)
The spatial distribution of calcium signals in squid presynaptic terminals.
J Physiol (Lond)
472:573-593[Abstract/Free Full Text].
-
Swandulla D,
Hans M,
Zipser K,
Augustine GJ
(1991)
Role of residual calcium in synaptic depression and posttetanic potentiation: fast and slow calcium signalling in nerve terminals.
Neuron
7:915-926[Web of Science][Medline].
-
Tank DW,
Regehr WG,
Delaney KR
(1995)
A quantitative analysis of presynaptic calcium dynamics that contribute to short-term enhancement.
J Neurosci
15:7940-7952[Abstract].
-
Wickelgren WO
(1977)
Physiological and anatomical characteristics of reticulospinal neurones in lamprey.
J Physiol (Lond)
270:89-114[Abstract/Free Full Text].
Copyright © 1999 Society for Neuroscience 0270-6474/99/191193-13$05.00/0
This article has been cited by other articles:
|
|
|
|
 
E. S. Hill and P. S. Katz
Role of Membrane Potential in Calcium Signaling During Rhythmic Bursting in Tritonia Swim Interneurons
J Neurophysiol,
March 1, 2007;
97(3):
2204 - 2214.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. Yang, G. L. Woodhall, and R. S. G. Jones
Tonic Facilitation of Glutamate Release by Presynaptic NR2B-Containing NMDA Receptors Is Increased in the Entorhinal Cortex of Chronically Epileptic Rats
J. Neurosci.,
January 11, 2006;
26(2):
406 - 410.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
S. Rossignol, R. Dubuc, and J.-P. Gossard
Dynamic Sensorimotor Interactions in Locomotion
Physiol Rev,
January 1, 2006;
86(1):
89 - 154.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
H. Photowala, R. Freed, and S. Alford
Location and function of vesicle clusters, active zones and Ca2+ channels in the lamprey presynaptic terminal
J. Physiol.,
November 15, 2005;
569(1):
119 - 135.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
S. Bevan and D. Parker
Metaplastic Facilitation and Ultrastructural Changes in Synaptic Properties Are Associated with Long-Term Modulation of the Lamprey Locomotor Network
J. Neurosci.,
October 20, 2004;
24(42):
9458 - 9468.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
R. Bardoni, C. Torsney, C.-K. Tong, M. Prandini, and A. B. MacDermott
Presynaptic NMDA Receptors Modulate Glutamate Release from Primary Sensory Neurons in Rat Spinal Cord Dorsal Horn
J. Neurosci.,
March 17, 2004;
24(11):
2774 - 2781.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. Parker
Activity-Dependent Feedforward Inhibition Modulates Synaptic Transmission in a Spinal Locomotor Network
J. Neurosci.,
December 3, 2003;
23(35):
11085 - 11093.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. Takahashi and S. Alford
The Requirement of Presynaptic Metabotropic Glutamate Receptors for the Maintenance of Locomotion
J. Neurosci.,
May 1, 2002;
22(9):
3692 - 3699.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
G. Woodhall, D. I. Evans, M. O. Cunningham, and R. S. G. Jones
NR2B-Containing NMDA Autoreceptors at Synapses on Entorhinal Cortical Neurons
J Neurophysiol,
October 1, 2001;
86(4):
1644 - 1651.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. Takahashi, R. Freed, T. Blackmer, and S. Alford
Calcium influx-independent depression of transmitter release by 5-HT at lamprey spinal cord synapses
J. Physiol.,
April 15, 2001;
532(2):
323 - 336.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
P. Saltiel, K. Wyler-Duda, A. D'Avella, M. C. Tresch, and E. Bizzi
Muscle Synergies Encoded Within the Spinal Cord: Evidence From Focal Intraspinal NMDA Iontophoresis in the Frog
J Neurophysiol,
February 1, 2001;
85(2):
605 - 619.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. Casado, S. Dieudonné, and P. Ascher
Presynaptic N-methyl-D-aspartate receptors at the parallel fiber-Purkinje cell synapse
PNAS,
September 29, 2000;
(2000)
200354297.
[Abstract]
[Full Text]
|
|
|
|
|
|
|
N. E. Schwartz and S. Alford
Physiological Activation of Presynaptic Metabotropic Glutamate Receptors Increases Intracellular Calcium and Glutamate Release
J Neurophysiol,
July 1, 2000;
84(1):
415 - 427.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
K. P. S. J. Murphy, R. J. Carter, L. A. Lione, L. Mangiarini, A. Mahal, G. P. Bates, S. B. Dunnett, and A. J. Morton
Abnormal Synaptic Plasticity and Impaired Spatial Cognition in Mice Transgenic for Exon 1 of the Human Huntington's Disease Mutation
J. Neurosci.,
July 1, 2000;
20(13):
5115 - 5123.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
P. Kloppenburg, W. R. Zipfel, W. W. Webb, and R. M. Harris-Warrick
Highly Localized Ca2+ Accumulation Revealed by Multiphoton Microscopy in an Identified Motoneuron and Its Modulation by Dopamine
J. Neurosci.,
April 1, 2000;
20(7):
2523 - 2533.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. Casado, S. Dieudonne, and P. Ascher
Presynaptic N-methyl-D-aspartate receptors at the parallel fiber-Purkinje cell synapse
PNAS,
October 10, 2000;
97(21):
11593 - 11597.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
Top
Abstract
Introduction
![[<UP>Ca<SUP>2+</SUP></UP>]−K<SUB>d</SUB> <FR><NU>(F−F<SUB><UP>min</UP></SUB>)</NU><DE>(F<SUB><UP>max</UP></SUB>−F)</DE></FR>.](/content/vol19/issue1/fulltext/193/img001.gif)
